Power Amplifier Longevity

These are typically the factors that reduce any electronic device’s life:

1) Inrush current at turn on.
2) Temperature.
3) AC voltage spikes which causes the maximum allowable voltages to exceed for various components.

Sooooo.. pick an amp which:

1) Has an inrush current limiter (lights do not dim when you switch your device on).
2) Runs cool.
3) Components in it are rated at twice the maximum rated operating voltages.

Pass labs seems to be one of the the few exceptions where most run pretty hot (class A) and yet still are very reliable. I can only guess that they must be using absolutely superb components inside (like 105 degree C capacitors) , which may also explain their prices.

@jollygreenaudiophile...

I guess you did not read the last part of my comment mentioning Pass Labs amps. Please read it again :-)

As for inrush current limiters, MOST devices sold today do **NOT** have such current limiters. Also, I am not quite sure how you got "fuses" lumped into the same category as a current limiting device but there is absolutely no correlation between them.

Any device which uses transformers (especially toroidals over a few hundred VAs), and electrolytic capacitors (such as the large filters on most power amps) present an almost SHORT CIRCUIT to the mains within the first cycle of the AC voltage, which lasts about 1/120th of a second. This is a very long time in the electronics world. Until the magnetic field is established on the transformer, the ONLY limit is the simple PURE wire resistance of the transformer. This can be in the very low ohms for large transformers (1 or 2 ohms max). We are talking in excess of 80 - 100 amps of inrush for 1/60th of a second. That is why lights dim for a second when some amps are switched on. In addition, on the transformer secondary side, the filter caps are a also virtual SHORT CIRCUIT as soon as power is applied. This means your rectifier is presented with a short circuit as soon as power is switched on. Again we are talking 100s of amps instantaneously until the caps start charging. Those currents, despite being on only about tenths of a second present a tremendous overload to the components, especially the rectifier and the capacitor itself.

Check out the circuit here from Nelson himself:
https://www.diyaudio.com/forums/diyaudio-com-articles/154777-burning-amplifier-ba-1-a.html

The items marked as THx (resistors with arrows thru them) are called Thermistors or NTCs (Negative temperature coefficient) meaning their resistance DROPS as temp INCREASES. Their purpose is to limit the inrush current.

Few commercial amps employ them as of today.

The point about doubling the voltage on components is simply for over-engineering. There is NEVER any harm in doing it except simply their expense. In addition, higher voltage capacitors have MUCH HIGHER ripple current ratings (check out digikey and search for them, look for low frequency ripple current ratings and ESR) which in the long run always make the components more reliable and less prone to spikes.

These things ALL add up, bit by bit, and increase the lifetime of an amp to 30 years from maybe 15....

Show me one piece of information in the following that indicates reduction of capacitance with its voltage rating:

https://content.kemet.com/datasheets/KEM_A4075_ALS70_71.pdf

Remember I am talking max voltage RATING not the voltage APPLIED.

Doubling the voltage will NOT halve the capacitance.  Where did you get that from ?

Show me ONE graph that shows this.

Hey atmasphere, good to hear your equipment has them.  Almost all transformers rated above about 300 VAs should have them but most do not.  Expensive gear is obviously better designed and it does.
Thanks

heaudio, I am sorry but you are not making any sense in your capacitance stuff.  Noone is APPLYING the higher voltage.  The voltage is FIXED at the rails.  The ONLY time it moves UP is when there is a surge.  The overvoltage RATING of the capacitor simply is able to absorb the extra max voltage which spikes so it does not blow the cap.  That is why I prefer the voltage rating to be high so that it indeed CAN absorb spikes.

There will be NO change whatsoever in the capacitance of a cap becoz of its voltage rating or by changing the applied voltage to it (unless temperature becomes an issue).

And the thermistors are NTC btw: 
https://www.newark.com/amphenol-advanced-sensors/cl-60/ntc-inrush-current-limiter-10/dp/81F3390

They have to be since they must settle to a LOW value with HIGHER temps otherwise you would never be able get more than quarter of the power of the amp since the current would be limited ALL the time.
A better (but more expensive) way of doing this would be using relays with a timer of course.

Quoted from United Chemicon:

(a) Operating Voltage

When in service at voltages equal to or below the rated value, the life of electrolytic capacitors is affected less by applied voltage than by operating temperature. Figures 7, 8 and 9 show life test results with various reduced voltages applied. The curves show that the life of the capacitor has not been significantly increased by a reduction in voltage. This is due to the use of proper forming voltages to minimize gas generation and leakage current. From this we can say that when capacitors are used at or below their rated voltage, the acceleration factor AV is equal to 1.

My point in choosing a capacitor to have a max voltage higher than normal is so that it handles the voltage spikes that are likely to happen.  If the AC mains was perfect, it would not be a problem.  But if any AC machinery or any kind of inductive load is running near the outlet to which your device is connected, if the devices do not have excellent filters, they will cause spikes on the AC line when powered off.  So, a cap with a higher voltage rating will favor better than one which does not.

I am not trying to increase the life of the capacitor by choosing a higher rating.  I am making it more resilient to possible spikes over the long term.  That is the essence of my argument.